Access control system and control method

Abstract

To minimize the occurrence of malfunctions caused by narrowing the effective recognition range of a first camera and a second camera when installing the cameras at a first gate device and a second gate device forming a single-person passage lane.SOLUTION: Face authentication processing is performed using face images captured by a camera installed on a gate device forming a single-person passage lane and registered face images, and gate control processing is performed using the result of detecting a person by a sensor installed on the gate device. In gate control processing, an open instruction is sent to the first gate device in a case 1 where a successful authentication result matching the registered face images with the face images captured by a first camera along with a person detection result from a first sensor are obtained, and in a case 2 where a successful authentication result matching the registered face images with the face images captured by a second camera along with a person detection result from the first sensor are obtained, but no person detection result is obtained from the second sensor.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present disclosure relates to a system and method for managing entry into and exit from a space requiring personal authentication. [Background technology]

[0002] Japanese Patent Application Laid-Open Publication No. 2019-053675 discloses a system that performs facial recognition of people passing through first and second gate devices arranged side by side using first and second cameras installed in the gate devices. This conventional system compares first and second captured images acquired at the same time by the first and second cameras. Furthermore, if a person's facial image is included in the edge area of ​​the first captured image and a person's facial image is included in the central area of ​​the second captured image, the conventional system determines that the people in these facial images are the same person. The system then excludes the person's facial image included in the edge area of ​​the first captured image from the target of facial recognition.

[0003] In addition to JP 2019-053675 A, JP 2017-148987 A can be cited as examples of documents that show the state of the art in the technical field related to the present disclosure. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2019-053675 [Patent Document 2] Japanese Patent Application Laid-Open No. 2017-148987 Summary of the Invention [Problem to be solved by the invention]

[0005] Consider a case where the first and second gate devices each form a lane for one person, and the effective recognition ranges of the first and second cameras are set so that they do not overlap. In this case, conventional systems do not require processes such as comparing the first and second captured images or determining whether they are the same person. Therefore, it is possible to recognize a person passing through the first gate device based on the first captured image, and a person passing through the second gate device based on the second captured image. However, this narrows the effective recognition range of each camera. Therefore, there is a possibility that a problem may occur, such as a person approaching the first gate device from the blind spot of the first camera being unable to pass through the first gate device.

[0006] One object of the present disclosure is to provide a technology that can reduce the occurrence of problems associated with narrowing the effective recognition range of a first and second camera installed at a first and second gate device, respectively, that forms a lane for single-passage. [Means for solving the problem]

[0007] A first aspect of the present disclosure is an entrance / exit management system having the following features. The management system includes a gate device, a camera, a sensor, and a processing device. The gate device forms a lane for single-passage. The gate device opens the lane in response to an open command and closes the lane in response to a close command. The camera is provided in the gate device. The camera acquires a facial image of a person entering the lane. The processing device performs a facial recognition process using a registered facial image and a facial image acquired by the camera, and a gate control process using the result of the facial recognition process and the result of person detection by the sensor. The gate device includes a first gate device forming a first lane and a second gate device arranged parallel to the first gate device and forming a second lane. The cameras include a first camera provided in the first gate device and a second camera provided in the second gate device. The sensors include a first sensor provided in the first gate device and a second sensor provided in the second gate device. The orientation direction of the first camera is the same as that of the second camera. The effective recognition range of the first camera does not overlap with that of the second camera. In the gate control process, the processing device transmits the open instruction to the first gate device in a first case where a successful authentication result is obtained in which the registered face image matches the face image acquired by the first camera and a person detection result is obtained by the first sensor, and also transmits the open instruction to the first gate device in a second case where a successful authentication result is obtained in which the registered face image matches the face image acquired by the second camera and a person detection result is obtained by the first sensor, but a person detection result is not obtained by the second sensor.

[0008] A second aspect of the present disclosure is an entrance / exit management method, which has the following characteristics. The management method includes the steps of performing face recognition processing using a face image acquired by a camera provided in a gate device that forms a lane for single-passage, opens the lane in accordance with an open command, and closes the lane in accordance with a close command, and a registered face image, and performing gate control processing using the result of the face recognition processing and the result of person detection by a sensor provided in the gate device. The gate device includes a first gate device forming a first lane and a second gate device arranged parallel to the first gate device and forming a second lane. The cameras include a first camera provided in the first gate device and a second camera provided in the second gate device. The sensors include a first sensor provided in the first gate device and a second sensor provided in the second gate device. The orientation direction of the first camera is the same as that of the second camera. The effective recognition range of the first camera does not overlap with that of the second camera. In the step of performing the gate control processing, in the first case where a successful authentication result is obtained that the registered face image matches the face image acquired by the first camera and a person detection result is obtained by the first sensor, the open instruction is sent to the first gate device, and in the second case where a successful authentication result is obtained that the registered face image matches the face image acquired by the second camera and a person detection result is obtained by the first sensor, but a person detection result is not obtained by the second sensor, the open instruction is also sent to the first gate device. [Effects of the Invention]

[0009] According to the present disclosure, an open instruction is sent to the first gate device not only in a first case where a successful authentication result is obtained in which the registered face image matches the face image acquired by the first camera and a person detection result is obtained by the first sensor, but also in a second case where a successful authentication result is obtained in which the registered face image matches the face image acquired by the second camera and a person detection result is obtained by the first sensor, but a person detection result is not obtained by the second sensor. Therefore, even if the effective recognition ranges of the first and second cameras are narrowed so that these ranges do not overlap, it is possible to allow a person approaching the first gate device from the recognition blind spot of the first camera to pass. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating a configuration example of an entrance / exit management system according to an embodiment of the present disclosure. [Figure 2] FIG. 10 is a diagram showing an example of the configuration of user reservation information. [Figure 3] FIG. 10 is a diagram showing an example of the configuration of user registration information. [Figure 4] FIG. 10 is a diagram illustrating an example of setting an effective recognition range of a camera. [Figure 5] 10 is a flowchart illustrating an example of information processing particularly related to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In each drawing, the same or corresponding parts are denoted by the same reference numerals, and the description thereof will be simplified or omitted.

[0012] 1.Configuration example 1 is a diagram illustrating a configuration example of an entrance / exit management system according to an embodiment of the present disclosure. In the example shown in FIG. 1, the entrance / exit management system includes three gate devices G1, G2, and G3, information processing devices 6 and 7, and a user terminal 5.

[0013] The gate devices G1 to G3 are installed at the entrances and exits of spaces ZN where personal authentication is required, such as event venues and residential areas. Examples of events held at event venues include any occasions, activities, competitions, etc. held during a specific time period. A residential area may be the site of one or more detached houses, or the site of one or more apartment buildings (e.g., condominiums). A residential area may also include various facilities other than residential buildings (e.g., commercial facilities, public facilities, educational facilities, medical and welfare facilities, etc.).

[0014] Gate devices G1 to G3 are arranged in parallel. Gate device G1 forms lane L1. Gate device G3 forms lane L2. Gate device G2 forms lane L1 or L2. One of lanes L1 and L2 is an example of a "first lane" in the present disclosure, and the other is an example of a "second lane" in the present disclosure. One of lanes L2 and L3 is an example of a "first lane" in the present disclosure, and the other is an example of a "second lane" in the present disclosure.

[0015] The gate devices G1 to G3 have the same basic configuration. Gate device G1 manages entry via lane L1. Gate device G3 manages exit via lane L2. Gate device G2 manages entry via lane L2 and exit via lane L1. The number of gate devices G2 to be installed is determined arbitrarily depending on the width of the entrance / exit. For example, if the entrance / exit is wide, two or more gate devices G2 are installed, and if the entrance / exit is narrow, the number of gate devices G2 is zero.

[0016] Hereinafter, unless the gate devices G1 to G3 are specifically described, they will be collectively referred to as "gate device G." Gate device G is a normally closed flapper type gate device. Gate device G comprises a gate device main body 1, a flapper 2, a camera 3, and a sensor 4. Gate device G uses camera 3 to perform facial recognition of people approaching gate device G. Gate device G also uses sensor 4 to detect people entering the lane of gate device G (lane L1 or L2).

[0017] Controlling the open / closed state of the flapper 2 based on a combination of the person's face authentication result and the person detection result is an example of a "gate control process" in the present disclosure. For example, if the face authentication result is "authentication successful" and a person detection result is obtained, the flapper 2 is set to the open state. If the face authentication result is "authentication failed" or if a person detection result is not obtained, the flapper 2 is set to the closed state. When the flapper 2 is set to the open state, the flapper 2 of the gate device G that is paired with this flapper 2 is also set to the open state, thereby opening the lane (lane L1 or L2) of the gate device G. On the other hand, when the flapper 2 is set to the closed state, this lane (lane L1 or L2) is closed. Note that the detection of a person is determined, for example, by the output of a sensor signal SNS from the sensor 4.

[0018] Person face authentication using the camera 3, person detection using the sensor 4, and control of the flapper 2 are performed by information processing devices 6 and 7. Each of the information processing devices 6 and 7 is a computer equipped with at least one processor and at least one storage device. The difference between the information processing device 6 and the information processing device 7 is that the former is a device that manages the gate device G, while the latter manages the entire entrance / exit control system. However, the information processing devices 6 and 7 may be configured from a single computer, or may be configured from three or more computers.

[0019] 2. User U's entry and exit In the embodiment, we consider the entry and exit of a user U who uses the space ZN into and from the space ZN. Facial authentication of a person using the camera 3 is performed by comparing a facial image IMG_HF (acquired facial image) acquired using the camera 3 with a facial image IMG_UF (registered facial image) of the user U that has been registered in advance.

[0020] The facial image IMG_HF is acquired by a facial acquisition processing unit 61 of the information processing device 6. The facial acquisition processing unit 61 is realized by, for example, a processor included in the information processing device 6. The imaging range of the camera 3 includes the lane of the gate device G in which the camera 3 is installed. Therefore, when a person approaches the gate device G, the image information IMG acquired by the camera 3 installed in the gate device G includes an image of the person. The facial acquisition processing unit 61 extracts the facial image IMG_HF of the person from the image of the person included in the image information IMG. In this way, the facial image IMG_HF is acquired. The facial acquisition processing unit 61 outputs the facial image IMG_HF to the face authentication processing unit 62.

[0021] The facial image IMG_UF is registered using the user terminal (e.g., a tablet, a smartphone, etc.) 5 of the user U. The facial image IMG_UF is included in reservation information RES transmitted from the user terminal 5 to a reservation processing unit 71 of the information processing device 7. The reservation processing unit 71 is realized, for example, by a processor possessed by the information processing device 7. When the reservation information RES is received from the user terminal 5, the reservation processing unit 71 generates registration information USR of the user U based on the reservation information RES and stores it in the database 73 of the information processing device 7.

[0022] The facial image IMG_HF and the facial image IMG_UF are matched by the facial authentication processing unit 62 of the information processing device 6 and the information management unit 72 of the information processing device 7. The facial authentication processing unit 62 is realized, for example, by a processor included in the information processing device 6. The information management unit 72 is realized, for example, by a processor included in the information processing device 7. When the facial image IMG_HF is input from the face acquisition processing unit 61, the facial authentication processing unit 62 queries the information management unit 72 to see if there is a facial image IMG_UF that matches this facial image IMG_HF.

[0023] When an inquiry is made by the face authentication processing unit 62, the information management unit 72 calculates the similarity S between any face image IMG_UF stored in the database 74 and the face image IMG_HF. If there is a face image IMG_UF whose similarity S is equal to or greater than a threshold value TH, the information management unit 72 determines that there is a face image IMG_UF that matches the face image IMG_HF, and sends answer information (authentication successful) to the face authentication processing unit 62. Otherwise, the information management unit 72 sends answer information (authentication failed) to the face authentication processing unit 62. The information management unit 72 periodically stores the face image IMG_UF in the database 74 based on the registration information USR stored in the database 73.

[0024] When the face authentication processing unit 62 receives the answer information from the information management unit 72, the face authentication processing unit 62 generates a face authentication result based on the answer information and transmits it to the passage management unit 63. The passage management unit 63 is realized, for example, by a processor included in the information processing device 6. When the face authentication processing unit 62 receives the face authentication result (authentication success) from the face authentication processing unit 62, the passage management unit 63 waits for input of a sensor signal SNS from the sensor 4 of the same gate device G (hereinafter also referred to as the "same gate") as the gate device G in which the camera 3 that acquired the face image IMG_HF is installed. Then, when the sensor signal SNS is input from the sensor 4 of the same gate, the passage management unit 63 generates an instruction INS to permit entry and exit of the user U and transmits it to the gate device G in which the camera 3 that acquired the face image IMG_HF is installed. As a result, the flapper 2 of the gate device G in which the camera 3 that acquired the face image IMG_HF is installed is controlled to an open state. The process for controlling the flapper 2 by the passage management unit 63 is an example of the "gate control process" of the present disclosure.

[0025] FIG. 2 is a diagram showing an example of the configuration of reservation information RES for user U. In the example shown in FIG. 2, the reservation information RES includes identification information UID, date and time information DAH, and a user image UIM. Examples of the identification information UID include attribute information of user U (e.g., name, gender, age). The date and time information DAH is information about the date and time period when user U will use space ZN. The date and time information DAH is, for example, information about the date and time period when user U wishes to use space ZN. User U may optionally input the date and time information DAH. In other words, the date and time information DAH is optional information. The user image UIM is an image related to user U. An example of the user image UIM is user U's facial image IMG_UF.

[0026] FIG. 3 is a diagram showing an example of the configuration of registration information USR for user U. In the example shown in FIG. 3, the registration information USR includes identification information UID, a user image UIM, and additional information ADD_U. The identification information UID, identification information AID, and user image UIM are common to the information that constitutes the reservation information RES. This is because the registration information USR is generated based on the reservation information RES. As explained in the explanation of FIG. 1, the registration information USR is generated by the reservation processing unit 71 shown in FIG. 1.

[0027] The information specific to FIG. 3 is date and time information DAH. The date and time information DAH is generated based on the reservation information RES or based on the date and time of an event scheduled to be held in space ZN. Information on events scheduled to be held in space ZN is generated by referencing space information ZNE stored in database 75. If space ZN is a residential area, the date and time information DAH may be generated based on the day of the week and time of day when user U passes through gate device G. The generation of date and time information DAH is performed by the information management unit 72 shown in FIG. 1.

[0028] The date and time information DAH is a parameter that adjusts the threshold value TH that is compared with the similarity S. For example, consider a case where an event is scheduled to be held in the space ZN. In this case, if the current time falls within the time period of the scheduled event, the threshold value TH can be set to a low value to alleviate congestion, or to a high value to strengthen security. As another example, consider a case where the date and time information DAH is included in the reservation information RES. In this case, the threshold value TH can be set to a low value to make it easier for the user U to pass through the gate device G.

[0029] 3. Features of the embodiment In this embodiment, we also consider a case where lanes L1 and L2 each form a lane for single-person passage. Furthermore, we consider a case where the effective recognition ranges of the two cameras 3 installed at two adjacent gate devices G are set so as not to overlap with each other. That is, we consider a case where the effective recognition range of the camera 3 at gate device G1 does not overlap with that of the camera 3 at gate device G2, and the effective recognition range of the camera 3 at gate device G2 does not overlap with that of the camera 3 at gate device G3.

[0030] Fig. 4 is a diagram illustrating an example of setting the effective recognition range of the camera 3 provided at each gate device. In the example shown in Fig. 4, the orientation direction D1 of the optical axis of the camera 3 at gate device G1 and the orientation direction D21 of the optical axis of the camera 3 at gate device G2 are the same. Furthermore, the orientation direction D22 of the optical axis of the camera 3 at gate device G2 and the orientation direction D3 of the optical axis of the camera 3 at gate device G3 are the same. In the example shown in Fig. 4, the imaging range of each camera 3 at gate devices G1 to G3 is narrowed.

[0031] By adjusting the pointing direction and imaging range of the camera 3 in this way, it is possible to set the effective recognition range R1 of the camera 3 of the gate device G1 so that it does not overlap with the effective recognition range R21 of the gate device G2. It is also possible to set the effective recognition range R22 of the camera 3 of the gate device G2 so that it does not overlap with the effective recognition range R3 of the gate device G3.

[0032] In the example shown in FIG. 4, the effective recognition range R1 includes the end 8 of the gate device G1 located in the direction in which lane L1 is formed, but does not include the end 8 of the gate device G2 located in the direction in which lane L2 is formed. The effective recognition range R21 includes the end 8 of the gate device G2 located in the direction in which lane L2 is formed, but does not include the end 8 of the gate device G1 located in the direction in which lane L1 is formed. The effective recognition range R22 includes the end 8 of the gate device G2 located in the direction in which lane L1 is formed, but does not include the end 8 of the gate device G3 located in the direction in which lane L2 is formed. The effective recognition range R3 includes the end 8 of the gate device G3 located in the direction in which lane L2 is formed, but does not include the end 8 of the gate device G2 located in the direction in which lane L1 is formed.

[0033] Incidentally, when lanes L1 and L2 each form a lane for single-passage and the effective recognition ranges of the cameras 3 of the two adjacent gate devices G do not overlap, a blind spot for recognition occurs. Therefore, a user U who approaches gate device G from the gate device G2 side and attempts to enter via lane L1 may experience "authentication failure" in the facial recognition using the camera 3 of gate device G1. Also, a user U who approaches gate device G from the gate device G1 side and attempts to exit via lane L2 may experience "authentication failure" in the facial recognition using the camera 3 of gate device G2.

[0034] Therefore, in the embodiment, when a face authentication result (authentication success) of the user U is obtained, an instruction INS to permit entry and exit of the user U is generated not only when a sensor signal SNS is input from the sensor 4 of the same gate, but also when a sensor signal SNS is input from a sensor 4 of a gate device G (hereinafter also referred to as "another gate") other than the gate device G where the camera 3 that acquired the face image IMG_HF is installed. This instruction INS is transmitted to the gate device G where the sensor 4 that output the sensor signal SNS is installed. This makes it possible to prevent problems caused by the above-mentioned blind spots in recognition.

[0035] In this way, in the embodiment, the user U can enter and exit through the same gate or a different gate. However, when a face authentication result (authentication success) of the user U is obtained, if a sensor signal SNS is input from both the sensor 4 of the same gate and the sensor 4 of the different gate, there is a possibility that a person other than the user U has been detected by the sensor 4 of one of the gate devices G.

[0036] Therefore, in the embodiment, priority is given to the input of the sensor signal SNS from the sensor 4 of the same gate. That is, when a face authentication result (authentication success) of the user U is obtained and a sensor signal SNS is input from the sensor 4 of the same gate, the user U is permitted to enter or exit via the lane of the same gate. This case corresponds to the "first case" of the present disclosure. Also, in the embodiment, when a face authentication result (authentication success) of the user U is obtained and a sensor signal SNS is not input from the sensor 4 of the same gate but a sensor signal SNS is input from the sensor 4 of a different gate, the user U is permitted to enter or exit via the lane of the different gate. This case corresponds to the "second case" of the present disclosure.

[0037] 4. Information processing example Fig. 5 is a flowchart showing an example of information processing particularly related to the embodiment. The routine shown in Fig. 5 is repeatedly executed at a predetermined cycle by cooperative processing by the processors of the information processing device 6 and the information processing device 7 shown in Fig. 1.

[0038] 5, first, image information IMG is acquired (step S11). The image information IMG is acquired from the camera 3. The image information IMG includes, for example, identification information of the camera 3 that acquired the image information IMG and time information when the image information IMG was acquired. In the processing of step S11, if the image information IMG includes an image of a person, a facial image IMG_HF of the person is extracted.

[0039] Following the process of step S11, it is determined whether or not a facial image IMG_HF has been extracted (step S12). If the determination result of step S12 is negative, the process shown in Fig. 5 ends. If the determination result of step S12 is positive, facial recognition processing is performed (step S13).

[0040] In the face recognition process, it is determined whether there is a face image IMG_UF whose similarity S with the face image IMG_HF is higher than a threshold value TH. The similarity S is calculated using the face image IMG_HF extracted in the process of step S11 and an arbitrary face image IMG_UF stored in the database 74. If the result of this determination is positive, it is determined that the face images IMG_HF and IMG_UF match, and authentication success information is output. If not, authentication failure information is output. A fixed value or a variable value may be used for the threshold value TH. In the latter case, the calculation is performed by referring to the date and time information DAH stored in the database 74 or the spatial information ZNE stored in the database 75.

[0041] Following the process of step S13, it is determined whether or not authentication success information has been output (step S14). If the determination result of step S14 is negative, that is, if authentication failure information has been output, an instruction INS not to permit entry or exit is output (step S19). In this case, the flapper 2 is controlled to the closed state in accordance with the instruction INS.

[0042] If the determination result of step S14 is positive, the process of step S15 is performed. In the process of step S15, it is determined whether or not a sensor signal SNS is input from the sensor 4 of the same gate. If the determination result of step S15 is positive, an instruction INS to permit entry and exit is output (step S16). In this case, the flapper 2 is controlled to the open state in accordance with the instruction INS.

[0043] If the determination result in step S15 is negative, the process of step S17 is performed. In the process of step S17, it is determined whether or not a sensor signal SNS is input from a sensor 4 of another gate. If the determination result in step S17 is positive, the process of step S16 is performed. In other words, if a sensor signal SNS is not input from a sensor 4 of the same gate but a sensor signal SNS is input from a sensor 4 of another gate, an instruction INS to permit entry or exit is output.

[0044] If the determination result in step S17 is negative, it is determined whether the elapsed time TL has exceeded the permissible time TP (step S18). The elapsed time TL is calculated, for example, from the timing when a positive result is output in the processing of step S14, or from the timing when the processing of step S15 starts. The permissible time TP is the time during which the positive result from the processing of step S14 is maintained. The permissible time TP is set appropriately depending on the installation location of the gate device G, the content of the event, etc.

[0045] If the determination result in step S18 is positive, it is determined that the user U who was authenticated in the face authentication process in step S13 has no intention of entering or exiting. Therefore, in this case, the process of step S19 is performed. If the determination result in step S18 is negative, the process returns to the process of step S15 to determine whether a sensor signal SNS has been input from the sensor 4 at the same gate or a different gate. [Explanation of symbols]

[0046] 1...Gate device main body, 2...Flapper, 3...Camera, 4...Sensor, 5...User terminal, 6, 7...Information processing device, 8...End portion, 61...Face acquisition processing unit, 62...Face authentication processing unit, 63...Passage management unit, 71...Reservation processing unit, 72...Information management unit, 73, 74, 75...Database, U...User, S...Similarity, D1, D21, D22, D3...Direction, G1, G2, G3...Gate device, L1, L2...Lane, R1, R21, R22, R3...Effective recognition range, LT...Elapsed time, PT...Allowed time, ZN...Space, UID...Identification information, DAH...Date and time information, IMG...Image information, UIM...User image, IMG_UF...Face image, INS...Instruction, RES...Reservation information, USR...Registration information, ZNE...Space information, ADD_U...Additional information

Claims

1. a gate device that forms a lane for one person to pass through, opens the lane in response to an open command, and closes the lane in response to a close command; a camera provided in the gate device for capturing a facial image of a person entering the lane; a sensor provided in the gate device for detecting entry of a person into the lane; a processing device that performs a face authentication process using a registered face image and a face image acquired by the camera, and a gate control process using a result of the face authentication process and a result of person detection by the sensor; Equipped with the gate device includes a first gate device forming a first lane and a second gate device arranged parallel to the first gate device and forming a second lane; the cameras include a first camera provided in the first gate device and a second camera provided in the second gate device; the sensors include a first sensor provided in the first gate device and a second sensor provided in the second gate device; The orientation of the first camera is the same as that of the second camera, The effective recognition range of the first camera does not overlap with that of the second camera, In the gate control process, the processing device In a first case where a successful authentication result is obtained that the registered face image matches the face image acquired by the first camera and a person is detected by the first sensor, the open instruction is sent to the first gate device; The opening instruction is also transmitted to the first gate device in a second case where an authentication success result is obtained in which the registered face image matches the face image acquired by the second camera, a person detection result by the first sensor is obtained, and a person detection result by the second sensor is not obtained. An entrance / exit management system characterized by the above.

2. In the face authentication process, the processing device When a result of the face authentication process indicates that the registered face image matches the face image acquired by the first camera, the result is retained for a predetermined time; The first case includes a case where a result of detecting a person by the first sensor is obtained while a successful authentication result in which the registered face image matches the face image acquired by the first camera is being held.

2. The entrance / exit management system according to claim 1.

3. In the face authentication process, the processing device If a result of the face authentication process indicates that the registered face image matches the face image acquired by the second camera, the result is held for a predetermined time; The second case includes a case where, while a successful authentication result is being held in which the registered face image matches the face image acquired by the second camera, a person detection result is obtained by the first sensor and a person detection result is not obtained by the second sensor.

2. The entrance / exit management system according to claim 1.

4. an effective recognition range of the first camera includes an end of the first gate device located in the formation direction of the first lane, and does not include an end of the second gate device located in the formation direction of the second lane; The effective recognition range of the second camera includes an end of the second gate device and does not include an end of the first gate device.

4. The entrance / exit management system according to claim 1, wherein the entrance / exit management system is a system for managing entrance / exit information.

5. a step of performing a facial recognition process using a facial image acquired by a camera provided in a gate device that forms a lane for one person to pass through, opens the lane in response to an open command, and closes the lane in response to a close command, and a registered facial image; performing a gate control process using a result of the face authentication process and a result of detecting a person by a sensor provided in the gate device; Including, the gate device includes a first gate device forming a first lane and a second gate device arranged parallel to the first gate device and forming a second lane; the cameras include a first camera provided in the first gate device and a second camera provided in the second gate device; the sensors include a first sensor provided in the first gate device and a second sensor provided in the second gate device; The orientation of the first camera is the same as that of the second camera, The effective recognition range of the first camera does not overlap with that of the second camera, In the step of performing the gate control process, in a first case where a successful authentication result is obtained in which the registered face image matches the face image acquired by the first camera, and a person is detected by the first sensor, the open instruction is transmitted to the first gate device; The open instruction is also transmitted to the first gate device in a second case where an authentication success result is obtained in which the registered face image matches the face image acquired by the second camera, a person detection result by the first sensor is obtained, and a person detection result by the second sensor is not obtained. The entrance and exit management method is characterized by the following.

Images